We have developed germline and conditional models for menin knockout in the mouse. These have shown a hyperplastic precursor phase that is striking in the pancreatic islets before overgrowth of a clonal tumor. The precursor phase could be an appropriate drug target before development of irreversible clonal steps. We have identified menin interaction with the nuclear transcription factors junD, NF-kappaB, RPA2, and a myosin (NMMHC IIA), and other interactions are being sought. Other groups have found menin bound to MLL in a COMPASS-like complex. We have found that FBP1 is a component of this complex. These studies should help identify its biochemical activities. Menin inactivation shifts junD from growth suppressor to growth promoter, representing the first plausible pathway for menin tumorigenesis. Another type of pathway arises from our observations that menin binds directly or indirectly to hundreds or even thousands of sites in chromatin. Studies by others in rodents and man have revealed that an MEN1-like state can arise rarely from heterozygous mutation of p27, a cyclin dependent kinase inhibitor (CDKI). Germline mutation of p27 is about 1/100 (1%) the frequency of MEN1 mutation in MEN1-like probands. Recently we found even more rare mutations for MEN1-like states in other CDKIs (p14, p18, p21). This similar phenotype suggestes that menin and CDKIs are nearby in one common pathway towards development of endocrine tumors.